Review Jorbox Cards

• Alcohol: Beer, wine, vodka, getting drunk, hangovers, bars.
• Explicit Intimacy: Sex, nudity, bedroom secrets, "sleeping together", physical arousal.
• Dating/Romance (Western Style): Boyfriends/girlfriends living together, casual dating. (Courtship/Rishta/Tea is acceptable).
• Drugs: Hard drugs (cocaine, heroin).
• Vulgarity: Extreme profanity or bathroom humor that crosses the line of decency.
• Religious Desecration:
  • NO making fun of religious rituals (Salah, Khutbah, Wudu).
  • NO trivializing theological concepts (Prophets, Angels, Qayamat, Soul).
  • NO placing religious terms in "silly" contexts (e.g., "I trade my soul/iman for...").

3. Cultural Adaptation Guidelines

If a card is clean but "boring" or "too Western":

• Replace Western Pop Culture:
  • David Blaine -> "A Jinn" or "The Magician"
  • Dr. Phil -> "Aamir Liaquat" or "The Morning Show"
  • Florida Man -> "Lahori Man" or "Pindi Boy"
• Replace Western Concepts:
  • Prom/Homecoming -> "Mehndi/Valima"
  • Bacon/Ham -> "Nihari/Paye"
  • Dollars/401k -> "Rupees/Committee"

4. How to Remove Cards

A. Manual Block Review (Preferred for Quality)

1. Read: Use view_file to read a block of 100-500 lines.
2. Identify: Note line numbers of problematic cards.
3. Edit: Use multi_replace_file_content to either:
   • Delete: Replace the content with an empty string or remove the line (if tool supports line deletion, otherwise empty string and cleanup later).
   • Modify: Reword the card to be culturally fit (e.g., change "Bar" to "Chai Dhaba").

B. Automated Scripting (Preferred for Bulk Cleaning)

1. Create a Python script to scan for specific keywords.
2. Use lists of known "bad" words (e.g., ["vodka", "bacon", "dating app"]).
3. Read the CSV, filter out rows containing these words, and write back to the file.
4. Always backup the file before mass-deleting.

5. Example Workflow

# Example logic for a quick filter script
bad_words = ['beer', 'wine', 'dating']
new_rows = []
for row in reader:
    if not any(word in row[0].lower() for word in bad_words):
        new_rows.append(row)

6. Verification

After edit/removal:

1. Check the total line count (wc -l public/data/game5.csv).
2. Read a random sample to ensure data integrity (lines didn't shift or break CSV format).